The locus coeruleus in schizophrenia: a postmortem study of noradrenergic neurones

Despite evidence for an abnormality of noradrenergic function in schizophrenia, it remains unclear whether the number of noradrenergic neurones is normal in patients with the disorder. In postmortem, formalin-fixed tissue from 15 schizophrenic patients and 18 controls matched for age and gender, we made estimates of the number and size of tyrosine-hydroxylase-immunoreactive cells in the locus coeruleus (LC). No significant difference was detected between these groups in the cross-sectional area or diameter of immunoreactive cell profiles. Profile number was not significantly affected by gender, side of the brainstem (left or right), postmortem interval or time in formalin; however, the levels of immunoreaction product (optical density) correlated significantly with our profile counts, which were lower on average in the schizophrenic group. When optical density was included as a covariate in our comparison (a repeated-measures analysis of variance) of schizophrenic and control cases, we found no difference between these groups in the number of neurones counted. An age-related decrease in profile number was detected, but no effect of age on our estimates of cell size was apparent. Our results highlight the importance of accounting for potential confounding variables, including variations in the quality of immunostaining, in investigations of this type. The findings presented here concur with previous studies suggesting that noradrenergic dysfunction in schizophrenia is not associated with an anatomical abnormality at the level of the LC.     

Biochemical mapping of noradrenergic nerves arising from the rat locus coeruleus

Norepinephrine (NE) concentration was measured in discrete brain regions of the rat following unilateral electrolytic lesions of the locus coeruleus, to determine the distribution of its noradrenergic neurons. Discrete brain nuclei and subdivisions were dissected from frozen sections, and norepinephrine was measured by a sensitive radio-isotopic assay.A significant reduction by 29–63% of control values in norepinephrine content was observed ipsilateral to the lesion in the following areas: all portions of the cerebral cortex examined (entorhinal, hippocampal, cingulate, parietal, and occipital areas), anterior half of the cerebellar cortex, hypothalamic periventricular and paraventricular nuclei, anterior ventral thalamic nucleus, ventral thalamic nucleus, and habenula. It appears that these regions receive unilateral innervation by axons from the locus coeruleus. In 3 regions (the medial geniculate body, inferior colliculus, and posterior half of the cerebellum), NE was reduced bilaterally in a pattern suggesting bilateral innervation from the locus coeruleus. Since no reduction in NE concentration occurred in the medial preoptic nucleus, nucleus interstitialis stria terminalis (ventralis), dorsomedial hypothalamic nucleus, or medial forebrain bundle, axons from noradrenergic neurons in the locus coeruleus do not appear to innervate these regions.The biochemical mapping of noradrenergic nerves from the locus coeruleus is discussed in relation to distribution studies based on the histofluorescence method.     

Induction of c-Fos immunoreactivity in the rat forebrain by conditioned and unconditioned aversive stimuli.

The protein product of the c-fos proto-oncogene was immunocytochemically localized in forebrain regions of adult male Lewis rats subjected to a physically aversive footshock stimulus or a Pavlovian-conditioned, non-aversive, auditory stimulus. Animals receiving the conditioned stimulus were first conditioned by repeatedly pairing electric footshock, the unconditioned stimulus (US), with an auditory cue, the conditioned stimulus (CS). These animals were later tested with the CS in the absence of the US, a procedure which, like footshock itself, suppresses immune function. In animals exposed to the conditioned or unconditioned stressor, c-Fos was strongly expressed in cells of the paraventricular nuclei (PVN) of the hypothalamus, some of which contain corticotropin-releasing hormone (CRH), and other forebrain areas directly associated with autonomic function, the ventral lateral septal nuclei (LSV), the medial amygdaloid nuclei (AME), the sensorimotor cortex, the basal ganglia and thalamic nuclei. Control animals exhibited very little or no c-Fos in the above areas. The identified forebrain nuclei can now be targeted for further study aimed at elucidating their role in stress-induced immune alteration.     

Excitatory effects of ACTH on noradrenergic neurons of the locus coeruleus in the rat

Responses of presumed noradrenaline-containing neurons in locus coeruleus (LC) to iontophoretically applied ACTH were determined. The full sequence ACTH1-39 and the partial sequence ACTH1-24 were excitatory on the majority of neurons tested. The partial sequence ACTH27-39 was without effect on most neurons. The presence of ACTH-containing fibers in LC together with the present results may suggest a controlling influence of the peptide on the noradrenaline system.     

Age-related decline in the activity of noradrenergic neurons of the rat locus coeruleus

Spontaneous neuronal activity of noradrenergic neurons located in the rat locus coeruleus was compared in 3-, 10- and 22-month-old rats. A statistically significant change in the frequency distribution pattern between the 3-month and the older rats was noticed. The mean firing rate was reduced by 21% and 25% in the 10- and 22-month-old rats, respectively.     

The locus coeruleus noradrenergic system in the rat brain studied by dual-probe microdialysis

A dual-probe microdialysis technique was applied to the locus coeruleus (LC) and prefrontal cortex (PFC) of the brain of conscious rats. One probe was implanted close to the LC and was used to apply receptor-specific compounds by retrograde microdialysis. The effects of the LC infusions were recorded by a sampling noradrenaline by a second probe that was implanted in the ipsilateral prefrontal cortex. Infusion of sodium channel blocker tetrodotoxin (1 μM: 90 min) into the LC decreased extracellular noradrenaline in the PFC to ≈ 20% of control values. Infusion of α₂-adrenoceptor agonist clonidine (100 μM, infused during 15 or 45 min) near to the LC, decreased extracellular noradrenaline in the PFC to 35 and 20% of controls, respectively. These results indicate that > 80% of the extracellular levels of noradrenaline in the PFC is derived from LC intervation, and confirms the importance of α₂-autoreceptors on noradrenergic neurons in the LC. Infusion of the cholinergic receptor agonist, carbachol (100 μM, 45 min) near to the LC increased extracellular noradrenaline in the PFC to ≈ 150% of controls. Infusions of the excitatory amino-acid agonists NMDA and kainate into the LC caused marked increases in extracellular noradrenaline in the PFC to 240 and 200% of controls, respectively. The experiments with clonidine, carbachol, NMDA and kainate were repeated in anesthetized rats. Clonidine and carbachol were similarly effective as in conscious animals but the effects of NMDA and kainate on extracellular noradrenaline in the PFC were clearly suppressed: 145 and 130% of controls, respectively. These results suggest that increased arousal or behavioural activation might have contributed to the increases in extracellular noradrenaline that was seen after infusion of the glutamate agonists. These results also provide evidence for localization of cholinergic-, NMDA-, non-NMDA-receptor on noradrenergic neurons in the LC. Finally it is concluded that dual-probe microdialysis is a useful method to further investigate the pharmacology of LC-noradrenergic neurons. Carbachol and clonidine are suitable tools for a rapid and reversible stimulation or inhibition, respectively, of noradrenergic LC neurons.     

Convulsive and non-convulsive ethanol withdrawal behaviour in rats with lesions of the noradrenergic locus coeruleus system

The ascending noradrenergic pathways from the locus coeruleus were lesioned bilaterally in 10 rats by intracerebral 6-hydroxydopamine injections. Ten rats were sham-operated. All animals were subjected to a 4-day ethanol intoxication period using intragastric intubation. Intoxication and withdrawal assessments were performed blindly. The 6-hydroxydopamine lesions did not appear to affect tolerance to ethanol. During withdrawal, however, lesioned animals showed minor, but statistically significant changes in scores of certain non-convulsive withdrawal signs, but incidence and intensity of spontaneous and audiogenic convulsive seizures were not different between the groups.     

Catecholaminergic systems and amygdala kindling development. Effects of bilateral lesions of substantia nigra dopaminergic or locus coeruleus noradrenergic neurones

The effects of the bilateral and selective destruction of substantia nigra (SN) dopaminergic or locus coeruleus (LC) noradrenergic neurones, consecutive to a local injection of 6-hydroxydopamine, were studied on the development of amygdala kindling. Immunohistochemical controls of lesions were performed using selective dopamine (DA) or norepinephrine (NE) antibodies. The results demonstrated that a massive destruction of SN pars compacta neurones did not modify the rate of kindling development. Conversely, the lesions of LC neurones (sparing lateral tegmental nuclei) markedly facilitated the development of amygdala kindling. This effect was related to the extent of NE denervation. Together, these results suggest that DA is not strongly involved in the development of kindling, and that the nigrostriatal output does not play a major role in the generalization of kindled seizures. In contrast, they confirm an inhibitory influence exerted by LC noradrenergic ascending pathways on the development of kindling.     

Agonist-induced internalization of corticotropin-releasing factor receptors in noradrenergic neurons of the rat locus coeruleus

Corticotropin-releasing factor (CRF) acts within the locus coeruleus (LC), to modulate activity of the LC-norepinephrine (NE) system. Combining molecular and cellular approaches, we demonstrate CRF receptor (CRFr) mRNA expression in Sprague-Dawley rat LC and provide the first in vivo evidence for agonist-induced internalization of CRFr. CRFr mRNA was detected in LC micropunches by RT-PCR. In dual labelling immunofluorescence studies, tyrosine hydroxylase (TH) containing neurons exhibited CRFr labelling. At the ultrastructural level, immunogold-silver labelling for CRFr was localized to the plasma membrane of TH-immunoperoxidase labelled dendrites. CRF (100 ng) injection into the LC produced a robust neuronal activation that peaked 10-15 min after injection and was maintained for the duration of the recording. This was associated with CRFr internalization in LC neurons that was apparent at 5 and 30 min after injection. By 5 min after injection the ratio of cytoplasmic to total dendritic CRFr-labelling was 0.81 +/- 0.01 in rats injected with CRF and 0.59 +/- 0.02 in rats injected with artificial cerebrospinal fluid (ACSF; P < 0.0001). Enhanced internalization of CRFr was maintained at 30 min after CRF injection, with the ratio being 0.86 +/- 0.02 for CRF-injected cases and 0.57 +/- 0.03 for ACSF-injected cases (P < 0.0001). Internalized CRFr was associated with early endosomes, indicative of degradation or recycling. Agonist-induced CRFr internalization in LC neurons may underlie acute desensitization to CRF or stress. This process may be a pivotal target by which stressors or pharmacological agents regulate the sensitivity of the LC-NE system to CRF and subsequent stressors.     

Corticotropin-releasing factor activates noradrenergic neurons of the locus coeruleus

Corticotropin-releasing factor (CRF) administered intraventricularly (0.5 nmol) was found to increase the discharge rates of locus coeruleus (LC) neurons in anesthetized rats. A similar effect on discharge rate was alos observed during direct application of CRF to LC neurons by pressure microapplication. Intraventricular administration of CRF-OH, previously demonstrated to be considerably less potent in releasing ACTH, did not alter LC firing rates. These data suggest that activation of these central noradrenergic neurons may constitute an integral part of the overall ‘stress response’ initiated by CRF release.     

Degeneration of noradrenergic fibres from the locus coeruleus causes tight-junction disorganisation in the rat brain

Although functional studies demonstrate that noradrenaline controls the permeability of the blood-brain barrier, it has never been determined whether this neurotransmitter regulates the tight junction (TJ) assembly that confers the barrier property to brain microvessels. We thus tested in rats the effect of pharmacological depletion of noradrenaline with the noradrenergic toxin DSP4 (5 mg/kg) on the expression of the TJ proteins zonula occludens-1 (ZO1) and occludin. The effectiveness of the lesion was confirmed by tyrosine hydroxylase immunoreactivity, which showed noradrenergic fibre reduction accompanied by debris and swollen fibres in DSP4-treated brains. Noradrenergic fibre degeneration caused: (i) gliosis; (ii) disappearance of TJ proteins in vascular cell-to-cell contacts (49.9 and 38.3% reductions for occludin and ZO1, respectively); (iii) a 49.2% decrease in total ZO1 protein, measured by Western blot analysis, parallel to a 39.5% decrease in ZO1 mRNA, measured by real-time PCR; and (iv) a relative increase in the beta occludin isoform (62.9%), with no change in total occludin protein or mRNA. The expression of endothelial brain antigen, a marker of a functionally competent brain endothelium, was also reduced. We conclude that damage to the ascending fibres from the locus coeruleus caused TJ disruption and gliosis, a sign of inflammation. These results imply that the locus coeruleus degeneration reported in Alzheimer's and Parkinson's diseases may contribute to these disorders by causing blood-brain barrier dysfunction. Whether the vascular damage is the result of impaired noradrenergic transmission or secondary to the inflammatory reaction remains to be determined.     

The biphasic effects of locus coeruleus noradrenergic activation on dendrodendritic inhibition in the rat olfactory bulb

Some forms of olfactory learning require intact noradrenergic terminals in the olfactory bulb that originate from the locus coeruleus. To clarify the action of noradrenergic inputs on the dendrodendritic interaction between mitral and granule cells in the rat olfactory bulb, we analyzed field potentials in the granule cell layer of the olfactory bulb evoked by paired-pulse stimulation of the lateral olfactory tract before and after the activation of the locus coeruleus. Locus coeruleus activation by glutamate injection in the vicinity of the nucleus changed only the test response without any effect on conditioning response. Paired-pulse inhibition measured from the ratio of test response amplitude to conditioning response amplitude was significantly depressed immediately after locus coeruleus activation. Conversely, 2 min later, paired-pulse inhibition was significantly potentiated. The significant potentiation of inhibition lasted for several minutes. The depression–potentiation sequence of paired-pulse inhibition was blocked by infusion of timolol, a β-antagonist, into the olfactory bulb, in a dose-dependent manner, but not by infusion of phentolamine, an α-antagonist. Infusion of isoproterenol, a β-agonist, into the bulb mimicked the depression of paired-pulse inhibition by locus coeruleus activation. These results suggest that glutamate activation of the locus coeruleus produces a depression–potentiation sequence in granule cell-mediated feedback inhibition onto mitral cells in the olfactory bulb through β-adrenergic receptors.     

Existence of glutamate in noradrenergic locus coeruleus neurons of rodents

This study distinguished three types of immunolabeled neurons in nucleus locus coeruleus (LC) of the rat and mouse: cells single labeled either for tyrosine hydroxylase-like immunoreactivity (TH-LI) or glutamate (Glu)-LI, and those double labeled for both antigens. Although the double labeled neurons tend to be located in the middle and ventral thirds of the rat LC nucleus, throughout its rostrocaudal extent, such feature was not apparent in the mouse. Quantitatively a majority of neurons cocontaining TH- and Glu-LI were commonly observed in the rat (62%) and mouse (77%) LC. Additional studies utilizing the combined retrograde and immunohistochemical labeling revealed that such a high incidence of coexistence of the TH- and Glu-LI was also represented by coeruleocortical neurons in the rat (69% and 75% of all ipsilateral and contralateral projection cells, respectively). A possible role of coeruleocortical neurons involvement in Glu- and norepinephrine-mediated target neuron dysfunction is discussed.     

Some electrophysiological and pharmacological properties of the cortical, noradrenergic projection of the locus coeruleus in the rat

The effect of repetitive stimulation of the locus coeruleus (LC) on the discharge rate of spontaneously active neurons of the visual, rostral and cingulate cortex was investigated in untreated and catecholamine-depleted rats under chloral hydrate anesthesia. In untreated animals, the inhibitory transsynaptic effects predominated over the excitatory ones. In catecholamine-depleted rats, the percentage of inhibited cells was significantly reducted in all areas. The vast majority of spontaneously active neurons in all cortical regions was depressed by microiontophoretically applied noradrenaline (NA). A few cells were resistant to NA; no excitatory effects were noticed on any cell. The transsynaptically mediated depression of the discharge rate of cells in all three cortical areas was reversibly antagonized by the iontophoretically administered beta-receptor blocking drug practolol. On the contrary the a-receptor blocking drugs piperoxane and WB4101 were ineffective in this respect. Thus, we tentatively conclude from these data that the NA-elicited depression of cells in the cortex is mediated by a receptor of the beta-type. Repetitive stimulation of the reticular formation elicited a desynchronizing effect on the EEG of chloral hydrate anaesthetized rats. LC stimulation, in contrast, hardly produced any modification of the EEG as judged by visual examination of the recordings.     

Increase in galanin gene expression in locus coeruleus neurones of the rat following reserpine treatment.

Changes in galanin mRNA content in locus coeruleus neurones of the rat were studied after reserpine treatment (10 mg/kg s.c.) using an in situ hybridization technique and an alkaline phosphatase labelled oligodeoxynucleotide probe. An increase in galanin mRNA content in locus coeruleus neurones was detected as early as 3 h after reserpine treatment when compared to vehicle injected rats. A maximal increase in mRNA content was detected at 3 days after reserpine treatment. This transient increase in galanin mRNA content had subsided by post treatment day 20. The response of locus coeruleus neurones to the single reserpine injection was heterogeneous; cells in the dorsal portion of the nucleus exhibiting a greater response than ventrally located cells.     

Cerebral blood flow and oxygen consumption in the rat brain after lesions of the noradrenergic locus coeruleus system

The effects of lesions of the locus coeruleus neuron system on cerebral metabolic rate for oxygen (CMR_O2) and blood flow (CBF) was evaluated in paralyzed and mechanically ventilated rats, using a¹³³xenon modification of the Kety-Schmidt inert gas technique. Bilateral electrothermic lesions of the nucleus locus coeruleus or bilateral 6-hydroxydopamine lesions of its ascending bindle caused no significant change in CBF or CMR_O2. The 6-hydroxydopamine lesions did not influence the CBF and CMR_O2 responses to hypercapnia and hypoxia.It is concluded that the locus coeruleus does not exert any resting tone on CBF and CMR_O2 and that no influence on the CBF and CMR_O2 responses to hypercapnia and hypoxia is mediated via its ascending projections.     

Dipropylacetate-induced quasi-morphine abstinence behaviour in the rat: Participation of the locus coeruleus system

The role of the noradrenergic system in quasi-morphine abstinence behaviour induced by di-n-propylacetate (DPA) in rats has been studied. Depletion of noradrenaline (NA), by treatment with FLA-63, decreased the number of body shakes and the extent of horizontal activity evoked by DPA. Almost total suppression of these symptoms was obtained by injection of 20 ng morphine bilaterally into the locus coeruleus (LC). Destruction of the LC system by electrolytic lesion of the LC or by injection of 6-hyrodroxy-dopamine into the dorsal bundle revealed that degeneration of the NA system by at least 80% appears to be necessary to decrease the number of body shakes. It is concluded that the noradrenergic LC system fulfills a modulatory role in quasi-morphine abstinence behaviour induced by DPA.     

Ipsapirone and 1-(2-pyrimidinyl)-piperazine increase rat locus coeruleus noradrenergic activity

The effects of systemically administered ipsapirone, an aryl-piperazine compound, and its major metabolite 1-(2-pyrimidinyl)-piperazine (1-PP), on locus coeruleus (LC) noradrenergic activity was investigated. On an equimolar basis both ipsapirone and 1-PP were approximately equipotent in increasing LC neuronal activity. However, pretreatment with 1-PP caused a significantly greater parallel shift to the right of the dose response curve for the inhibitory action of the LC alpha 2-receptor agonist clonidine compared to ipsapirone. Biochemically, pretreatment with SKF 525A, a compound which prevents the formation of 1-PP from ipsapirone, diminished the ipsapirone-induced increase in MOPEG-SO4 levels in the brainstem and cortex. These data, together with the findings that 1-PP is more potent than ipsapirone in displacing 3H-clonidine from cerebral cortical membranes, suggest that the parent drug influences LC neuronal activity via the action of I-PP on LC alpha 2-adrenoceptors.     

Activity of cat locus coeruleus noradrenergic neurons during the defense reaction

The single-unit activity of locus coeruleus noradrenergic (LC-NE) neurons was recorded in freely moving cats during naturally induced defense reactions. Defense reactions, consisting of arched back, piloerection, flattened ears and mydriasis, were elicited by exposing the cat either to a dog, or to a cat displaying aggressive behavior induced by electrical stimulation of the hypothalamus. LC-NE neurons were identified using previously established criteria, including suppression of firing during rapid eye movement (REM) sleep and in response to clonidine administration. Exposure to a dog evoked defense reactions and increased the tonic firing rate of LC-NE neurons (n = 8) from a baseline of approximately 0.9 spikes/s to approximately 2.5 spikes/s. Exposure to an aggressive cat evoked defense reactions that were qualitatively very similar to those produced by dog exposure, and elevated the tonic firing rate of LC-NE neurons (n = 8) from a baseline of approximately 1.0 spikes/s to approximately 2.5 spikes/s. In addition to these tonic elevations of activity, LC-NE neurons discharged in phasic bursts (as high as 10 spikes in a 500 ms period) in close association with specific threatening acts made by the dog or hypothalamically stimulated cat. The mere presence of a dog was sufficient to evoke tonic activation of LC-NE neurons, even in the absence of threatening advances by the dog, whereas exposure to a hypothalamically stimulated cat produced LC-NE neuronal activation only when the stimulated cat showed aggressive behavior. These results extend our previous work, which examined the response of LC-NE neurons to environmental and physiological stressors, into a more ethologically relevant domain, and suggest that LC-NE neuronal activation may play a role in the response to threatening or challenging situations.